An Improved Physically-Based Method for Geometric Restoration of Distorted Document Images
Identifieur interne : 000D31 ( Main/Exploration ); précédent : 000D30; suivant : 000D32An Improved Physically-Based Method for Geometric Restoration of Distorted Document Images
Auteurs : LI ZHANG [Singapour] ; YU ZHANG [Singapour] ; CHEW LIM TAN [Singapour]Source :
- IEEE transactions on pattern analysis and machine intelligence [ 0162-8828 ] ; 2008.
Descripteurs français
- Pascal (Inist)
- Intelligence artificielle, Analyse forme, Numérisation, Formation image, Forme géométrique, Contenu image, Reconnaissance caractère, Reconnaissance optique caractère, Restauration image, Traitement image, Gauchissement, Aplatissement, Métrique, Flexion, Transformation géométrique, Système masse ressort, Modélisation, Intégration numérique.
- Wicri :
- topic : Intelligence artificielle, Numérisation.
English descriptors
- KwdEn :
- Artificial intelligence, Bending, Character recognition, Digitizing, Flattening, Geometric transformation, Geometrical shape, Image content, Image processing, Image restoration, Imaging, Metric, Modeling, Numerical integration, Optical character recognition, Pattern analysis, Spring mass system, Warping.
Abstract
-In document digitization through camera-based systems, simple imaging setups often produce geometric distortions in the resultant 2D images because of the nonplanar geometric shapes of certain documents such as thick bound books, rolled, folded, or crumpled materials, etc. Previous work [1], [2], [3], [4] has demonstrated that arbitrary warped documents can be successfully restored by flattening a 3D scan of the document. These approaches use physically-based or relaxation-based techniques in their flattening process. While this has been demonstrated to be effective in rectifying the image content and improving OCR, these previous approaches have several limitations in terms of speed and stability. In this paper, we propose a distance-based penalty metric to replace the mass-spring model and introduce additional bending resistance and drag forces to improve the efficiency of the existing approaches. The use of Verlet integration and special plane collision handling schemes also help to achieve better stability without sacrificing efficiency. Experiments on various document images captured from books, brochures, and historical documents with arbitrary warpings have demonstrated large improvements over the existing approaches in terms of stability and efficiency.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000286
- to stream PascalFrancis, to step Curation: 000498
- to stream PascalFrancis, to step Checkpoint: 000253
- to stream Main, to step Merge: 000D43
- to stream Main, to step Curation: 000D31
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">An Improved Physically-Based Method for Geometric Restoration of Distorted Document Images</title>
<author><name sortKey="Li Zhang" sort="Li Zhang" uniqKey="Li Zhang" last="Li Zhang">LI ZHANG</name>
<affiliation wicri:level="4"><inist:fA14 i1="01"><s1>School of Computing, National University of Singapore, 3 Science Drive 2</s1>
<s2>Singapore 117543</s2>
<s3>SGP</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Singapour</country>
<wicri:noRegion>Singapore 117543</wicri:noRegion>
<orgName type="university">Université nationale de Singapour</orgName>
</affiliation>
</author>
<author><name sortKey="Yu Zhang" sort="Yu Zhang" uniqKey="Yu Zhang" last="Yu Zhang">YU ZHANG</name>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Institute of High Performance Computing, A*STAR, #01-01, 1 Science Park Road, The Capricorn, Science Park II</s1>
<s2>Singapore 117528</s2>
<s3>SGP</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Singapour</country>
<wicri:noRegion>Singapore 117528</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Chew Lim Tan" sort="Chew Lim Tan" uniqKey="Chew Lim Tan" last="Chew Lim Tan">CHEW LIM TAN</name>
<affiliation wicri:level="4"><inist:fA14 i1="01"><s1>School of Computing, National University of Singapore, 3 Science Drive 2</s1>
<s2>Singapore 117543</s2>
<s3>SGP</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Singapour</country>
<wicri:noRegion>Singapore 117543</wicri:noRegion>
<orgName type="university">Université nationale de Singapour</orgName>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">08-0175238</idno>
<date when="2008">2008</date>
<idno type="stanalyst">PASCAL 08-0175238 INIST</idno>
<idno type="RBID">Pascal:08-0175238</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000286</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000498</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000253</idno>
<idno type="wicri:doubleKey">0162-8828:2008:Li Zhang:an:improved:physically</idno>
<idno type="wicri:Area/Main/Merge">000D43</idno>
<idno type="wicri:Area/Main/Curation">000D31</idno>
<idno type="wicri:Area/Main/Exploration">000D31</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">An Improved Physically-Based Method for Geometric Restoration of Distorted Document Images</title>
<author><name sortKey="Li Zhang" sort="Li Zhang" uniqKey="Li Zhang" last="Li Zhang">LI ZHANG</name>
<affiliation wicri:level="4"><inist:fA14 i1="01"><s1>School of Computing, National University of Singapore, 3 Science Drive 2</s1>
<s2>Singapore 117543</s2>
<s3>SGP</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Singapour</country>
<wicri:noRegion>Singapore 117543</wicri:noRegion>
<orgName type="university">Université nationale de Singapour</orgName>
</affiliation>
</author>
<author><name sortKey="Yu Zhang" sort="Yu Zhang" uniqKey="Yu Zhang" last="Yu Zhang">YU ZHANG</name>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Institute of High Performance Computing, A*STAR, #01-01, 1 Science Park Road, The Capricorn, Science Park II</s1>
<s2>Singapore 117528</s2>
<s3>SGP</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Singapour</country>
<wicri:noRegion>Singapore 117528</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Chew Lim Tan" sort="Chew Lim Tan" uniqKey="Chew Lim Tan" last="Chew Lim Tan">CHEW LIM TAN</name>
<affiliation wicri:level="4"><inist:fA14 i1="01"><s1>School of Computing, National University of Singapore, 3 Science Drive 2</s1>
<s2>Singapore 117543</s2>
<s3>SGP</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Singapour</country>
<wicri:noRegion>Singapore 117543</wicri:noRegion>
<orgName type="university">Université nationale de Singapour</orgName>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title>
<title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title>
<idno type="ISSN">0162-8828</idno>
<imprint><date when="2008">2008</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title>
<title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title>
<idno type="ISSN">0162-8828</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Artificial intelligence</term>
<term>Bending</term>
<term>Character recognition</term>
<term>Digitizing</term>
<term>Flattening</term>
<term>Geometric transformation</term>
<term>Geometrical shape</term>
<term>Image content</term>
<term>Image processing</term>
<term>Image restoration</term>
<term>Imaging</term>
<term>Metric</term>
<term>Modeling</term>
<term>Numerical integration</term>
<term>Optical character recognition</term>
<term>Pattern analysis</term>
<term>Spring mass system</term>
<term>Warping</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Intelligence artificielle</term>
<term>Analyse forme</term>
<term>Numérisation</term>
<term>Formation image</term>
<term>Forme géométrique</term>
<term>Contenu image</term>
<term>Reconnaissance caractère</term>
<term>Reconnaissance optique caractère</term>
<term>Restauration image</term>
<term>Traitement image</term>
<term>Gauchissement</term>
<term>Aplatissement</term>
<term>Métrique</term>
<term>Flexion</term>
<term>Transformation géométrique</term>
<term>Système masse ressort</term>
<term>Modélisation</term>
<term>Intégration numérique</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Intelligence artificielle</term>
<term>Numérisation</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">-In document digitization through camera-based systems, simple imaging setups often produce geometric distortions in the resultant 2D images because of the nonplanar geometric shapes of certain documents such as thick bound books, rolled, folded, or crumpled materials, etc. Previous work [1], [2], [3], [4] has demonstrated that arbitrary warped documents can be successfully restored by flattening a 3D scan of the document. These approaches use physically-based or relaxation-based techniques in their flattening process. While this has been demonstrated to be effective in rectifying the image content and improving OCR, these previous approaches have several limitations in terms of speed and stability. In this paper, we propose a distance-based penalty metric to replace the mass-spring model and introduce additional bending resistance and drag forces to improve the efficiency of the existing approaches. The use of Verlet integration and special plane collision handling schemes also help to achieve better stability without sacrificing efficiency. Experiments on various document images captured from books, brochures, and historical documents with arbitrary warpings have demonstrated large improvements over the existing approaches in terms of stability and efficiency.</div>
</front>
</TEI>
<affiliations><list><country><li>Singapour</li>
</country>
<orgName><li>Université nationale de Singapour</li>
</orgName>
</list>
<tree><country name="Singapour"><noRegion><name sortKey="Li Zhang" sort="Li Zhang" uniqKey="Li Zhang" last="Li Zhang">LI ZHANG</name>
</noRegion>
<name sortKey="Chew Lim Tan" sort="Chew Lim Tan" uniqKey="Chew Lim Tan" last="Chew Lim Tan">CHEW LIM TAN</name>
<name sortKey="Yu Zhang" sort="Yu Zhang" uniqKey="Yu Zhang" last="Yu Zhang">YU ZHANG</name>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000D31 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000D31 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= Main |étape= Exploration |type= RBID |clé= Pascal:08-0175238 |texte= An Improved Physically-Based Method for Geometric Restoration of Distorted Document Images }}
This area was generated with Dilib version V0.6.32. |